Ideas regarding coeff. of coupling, racing sparks, and tuning procedures

["Tesla list" <tesla@xxxxxxxxxx>]

Original poster: "resonance" <resonance@xxxxxxxxxxxx>


With DRSSTC systems racing sparks begin occuring in the area k = 0.20 
to 0.22.  With solid state coils, running much lower potentials on 
the primary, usually 0.16 to 0.18 is a good design target.  This 
favors the solenoid type primary for solid state coils due to the 
lower primary voltages.
With the higher nst potentials on classic style coils, the coupling 
is usually reduced to the range of 0.08 to 0.12 as a max.  Racing 
sparks usually don't occur in this range or below if a flat spiral is 
used.  The higher pri voltages require greater separation of the 
pri-sec and lower coupling to prevent flashovers.  Solenoid type 
primaries can be used with caution to keep the k factor lower than 
0.12.  If a pole or PT single ended type xmfr is used another 
advantage is gained because the inside of the pri is grounded as well 
as the lower end of the sec coil so there is no HV between these two 
points.  With nst coils, center-tap grounded secs, the inside of the 
primary is usually at 6-7 kV with respect to the lowest turn on the 
sec coil so more stress is produced which can assist in the dreaded 
racing sparks if the coupling is too tight.
I've conducted a lot of experiments with raising and lower sec coils 
in the flat spiral designs by using a small feedscrew attached to a 
wired remote controlled gearmotor.  In small systems sometimes the 
greatest spark length occurs when the coupling is reduced much below 
what you think you should use.  Example, with our larger 18 inch dia. 
sec coilforms, the best spark occurs when the sec is elevated approx 
6.5 inches above the horizontal plane of the primary.  At 24 inch 
dia. sec coilforms, the best spark occured when the sec was elevated 
a full 9 inches above the primary!
We used a pri to sec coilform spacing of 1.5 inches for most of our 
experiments, ie, keeping this as a constant from which our designs were based.
Each experimenter should use some small 1/2 inch high wood blocks, 
and stacking them as necessary, to confirm the best configuration for 
the longest possible sparks.
I'm also a big fan of using Bob Svangren's (Seattle based coiler) 
segmented spark gap for taking measurements on the precise spark 
length of the sec coil's output.  Bobs system is a series of small 
horizontal nails hung below 3 inch long wooden dowels which are 
strung on a line of fiberglass monofilament fishing line.  I have one 
that is 25 feet long and enables you to tune any size coil with 
sparks lengths up to 10 ft.  It provides sec spark tuning accurate 
tuning down to 1 inch increments.  Simple to make and easy to 
use.  You can set your variac to around 60-70% output and see how far 
the spark will travel across the 1 inch spark gaps.  The last nail is 
connected to a ground wire and this "tap" is moved out segment by 
segment until the tuning point, ie, best spark length, is found for 
any type coil design.  Then adjustment is made to the primary and 
this is run again moving the tap in or out as required to get the 
longest spark.  By fiddling with this last tap point you can even see 
the effect of humidity, barametric pressure, temperature, and other 
factors on your coil's performance.
It's hard to describe so perhaps I will draw a sketch and submit it 
to Gerry R. for inclusion in the archives.
Dr. Resonance




> I was surprised to see that Terrry's primary is a pancake. The few 
> solid state coils I have seen in person had a deep primary conical 
> primary. I guess I'm asking what is the practical limit to the 
> coupling; and, is there any benefit to using a conical coil in a 
> IGBT driven coil?